Navigation Based on Fuzzy Cognitive Maps for Needs of Ubiquitous Robotics

Internet of Things (IoT) has become a phenomenon of last five years. There is a number of various applications and its potential is still growing. Robotics is also one of perspective areas for IoT but it has some specific requirements. The concept of the so-called intelligent space (IS) is especially convenient for robotic applications, which can profit from spatially distributed and mutually networked sensors. Firstly, it is not necessary to carry all necessary sensors on board and hereby robots become much cheaper and secondly, such a solution is able to better predict occurrence of specific events, too. A ubiquitous robot, known as Ubibot, is an issue of interconnecting of IoT and IS. In this paper we proposed our own concept of Ubibot, where for navigation purposes the so-called fuzzy cognitive map (FCM) was used. For adjusting FCM parameters our own adaptation method based on evolutionary algorithms was used. The proposed system was experimentally tested and obtained results were analysed.